Prepare input list

The input data must be in a named list, with the same names as the input list of the pipeline.

inputs(rnaseq_Sf)

List of length 5
names(5): in_seqfiles in_prefix in_genomeDir in_GTFfile in_runThreadN

For this pipeline, 5 inputs are required to be set, including in_seqfiles, in_prefix, in_genomeDir, in_GTFfile and in_runThreadN.

There are two different input lists, inputList and paramList. The inputList is used to define the inputs for each sample and will be submitted to different cluster nodes. The paramList is used to define parameters which are shared in all jobs.

Two following inputs should be listed in inputList.

• in_seqfiles: A list with the fastq files of each sample in each element. The names of the list are also required to be defined and can be the sample IDs. The length of the list will be the same as the number of samples, thus the list will be defined to inputList and assigned to different nodes for parallel computing.

• in_prefix The same as in_seqfiles, which define a list of sample IDs.

files <- normalizePath(list.files("data/", ".gz", full.names = TRUE))
files <- tapply(files, substring(basename(files), 1, 8), as.list)
inputList <- list(in_seqfiles = files,
                  in_prefix = as.list(names(files)))

These 3 parameter will be defined in paramList.

• in_genomeDir: The reference genome indexes for STAR.

• in_GTFfile: The gene annotation file in GTF format.

• in_runThreadN: The number of threads to run for each job.

paramList <- list(
    in_genomeDir = "/rpcc/bioinformatics/reference/STAR/GRCh38_100/",
    in_GTFfile = "/rpcc/bioinformatics/annotation/GENECODE/gencode.v25.annotation.gtf",
    in_runThreadN = 4
)

In some case, we need to modify the default arguments in some steps of a pipeline. For example,

arguments(rnaseq_Sf, "STAR")[[2]] <- "2"
head(arguments(rnaseq_Sf, "STAR"))

[[1]]
[1] "--outFilterMultimapNmax"

[[2]]
[1] "2"

[[3]]
[1] "--outSAMunmapped"

[[4]]
[1] "Within"

[[5]]
[1] "--outFilterMismatchNmax"

[[6]]
[1] "2"

Submit pipeline with SGE

The function runCWLBatch is used to submit the pipeline to cluster server. In addition to defining inputList and paramList, we need to define parallel parameters from the BiocParallel package. Here, we use “sge” to submit the jobs. The “sge” template is a bash script with some predefined parameters for “qsub”. The nodes queue name and number of slots/threads are variables from the template and can be assigned by the resources list.

res <- runCWLBatch(cwl = rnaseq_Sf, outdir = "output/",
                   inputList = inputList, paramList = paramList,
                   BPPARAM = BatchtoolsParam(
                       workers = lengths(inputList)[1], cluster = "sge",
                       template = "/rpcc/bioinformatics/sge_centos7.tmpl",
                       resources = list(queue = "centos7.q",
                                        threads = 4)))

That’s it! The fastqc files of each sample will be submitted to different nodes to run the whole pipeline automatically.

All the results have been collected to output directory of each sample. For example,

dir("output/HBR_Rep1")

 [1] "HBR_Rep1_ERCC-Mix2_Build37-ErccTranscripts-chr22.read1_fastqc.zip"
 [2] "HBR_Rep1_ERCC-Mix2_Build37-ErccTranscripts-chr22.read2_fastqc.zip"
 [3] "HBR_Rep1.featureCounts.txt"                                       
 [4] "HBR_Rep1.featureCounts.txt.summary"                               
 [5] "HBR_Rep1.geneBodyCoverage.curves.pdf"                             
 [6] "HBR_Rep1Aligned.sortedByCoord.out.bam"                            
 [7] "HBR_Rep1Aligned.sortedByCoord.out.bam.bai"                        
 [8] "HBR_Rep1Aligned.sortedByCoord.out.distribution.txt"               
 [9] "HBR_Rep1Aligned.sortedByCoord.out.featureCounts.txt"              
[10] "HBR_Rep1Aligned.sortedByCoord.out.featureCounts.txt.summary"      
[11] "HBR_Rep1Aligned.sortedByCoord.out.flagstat.txt"                   
[12] "HBR_Rep1Log.final.out"                                            
[13] "HBR_Rep1ReadsPerGene.out.tab"                                     

Build the reports

wflow_build()

Session information

sessionInfo()

R version 3.5.2 Patched (2018-12-31 r75935)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: CentOS release 6.4 (Final)

Matrix products: default
BLAS: /home/qhu/usr/R-3.5/lib64/R/lib/libRblas.so
LAPACK: /home/qhu/usr/R-3.5/lib64/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] parallel  stats4    stats     graphics  grDevices utils     datasets 
[8] methods   base     

other attached packages:
 [1] RcwlPipelines_0.0.0.9000 shiny_1.2.0             
 [3] jsonlite_1.6             BiocParallel_1.16.5     
 [5] BiocStyle_2.10.0         rmarkdown_1.11          
 [7] Rcwl_0.99.14             S4Vectors_0.20.1        
 [9] BiocGenerics_0.28.0      yaml_2.2.0              
[11] usethis_1.4.0            devtools_2.0.1          

loaded via a namespace (and not attached):
 [1] fs_1.2.6           RColorBrewer_1.1-2 progress_1.2.0    
 [4] rprojroot_1.3-2    tools_3.5.2        backports_1.1.3   
 [7] R6_2.3.0           lazyeval_0.2.1     colorspace_1.3-2  
[10] withr_2.1.2        tidyselect_0.2.5   gridExtra_2.3     
[13] prettyunits_1.0.2  processx_3.2.1     git2r_0.24.0      
[16] compiler_3.5.2     cli_1.0.1          xml2_1.2.0        
[19] influenceR_0.1.0   desc_1.2.0         bookdown_0.9      
[22] scales_1.0.0       checkmate_1.9.0    readr_1.3.1       
[25] callr_3.1.1        rappdirs_0.3.1     commonmark_1.7    
[28] stringr_1.3.1      digest_0.6.18      R.utils_2.7.0     
[31] pkgconfig_2.0.2    htmltools_0.3.6    sessioninfo_1.1.1 
[34] highr_0.7          htmlwidgets_1.3    rlang_0.3.1       
[37] rstudioapi_0.9.0   bindr_0.1.1        visNetwork_2.0.5  
[40] R.oo_1.22.0        dplyr_0.7.8        rgexf_0.15.3      
[43] magrittr_1.5       Rcpp_1.0.0         munsell_0.5.0     
[46] viridis_0.5.1      R.methodsS3_1.7.1  whisker_0.3-2     
[49] stringi_1.2.4      debugme_1.1.0      pkgbuild_1.0.2    
[52] plyr_1.8.4         grid_3.5.2         promises_1.0.1    
[55] crayon_1.3.4       hms_0.4.2          batchtools_0.9.11 
[58] knitr_1.21         ps_1.3.0           pillar_1.3.1      
[61] igraph_1.2.2       base64url_1.4      pkgload_1.0.2     
[64] XML_3.98-1.16      glue_1.3.0         evaluate_0.12     
[67] downloader_0.4     BiocManager_1.30.4 data.table_1.11.8 
[70] remotes_2.0.2      httpuv_1.4.5.1     testthat_2.0.1    
[73] gtable_0.2.0       purrr_0.2.5        tidyr_0.8.2       
[76] assertthat_0.2.0   ggplot2_3.1.0      xfun_0.4          
[79] mime_0.6           xtable_1.8-3       roxygen2_6.1.1    
[82] later_0.7.5        viridisLite_0.3.0  tibble_2.0.0      
[85] memoise_1.1.0      workflowr_1.1.1    bindrcpp_0.2.2    
[88] Rook_1.1-1         DiagrammeR_1.0.0   brew_1.0-6